Comparative study of multi-objective metaheuristics¶
HV quality indicator¶
| Â | MOEAD | MOEAD500 | MOEAD1000 | MOEAD2000 | MOEAD3000 | Friedman |
|---|---|---|---|---|---|---|
| DTLZ1 | 7.8615e-01 | 7.8737e-01 | 7.6941e-01 | 7.8777e-01 | 7.8786e-01 | + |
| DTLZ2 | 4.1607e-01 | 4.1632e-01 | 3.8609e-01 | 4.1588e-01 | 4.1660e-01 | + |
| DTLZ3 | 3.6567e-01 | 3.2125e-01 | 3.8138e-01 | 4.1426e-01 | 4.1668e-01 | + |
| DTLZ4 | 4.0890e-01 | 4.0895e-01 | 3.8887e-01 | 4.0898e-01 | 4.0958e-01 | + |
| DTLZ5 | 8.4730e-02 | 9.3595e-02 | 9.4051e-02 | 9.3527e-02 | 9.3474e-02 | + |
| DTLZ6 | 4.7492e-02 | 9.4469e-02 | 9.4967e-02 | 9.4403e-02 | 9.4344e-02 | + |
| DTLZ7 | 2.4756e-01 | 3.0747e-01 | 2.9524e-01 | 3.0745e-01 | 3.0722e-01 | + |
Wilcoxon rank-sum test table¶
| Â | MOEAD500 | MOEAD1000 | MOEAD2000 | MOEAD3000 |
|---|---|---|---|---|
| MOEAD | ===---- | ++=+--- | -=----- | ------- |
| MOEAD500 | ++-+--+ | ==-=++= | -=--++= | |
| MOEAD1000 | ----++- | ----++- | ||
| MOEAD2000 | =---++= |
Critical distance ranking¶
Boxplot + Pareto Fronts for each problem¶
